TWI335735B - Groupwise successive interference cancellation for block transmission with reception diversity - Google Patents

Description

B35735 IX. Description of the Invention: TECHNICAL FIELD OF THE INVENTION The present invention relates generally to wireless communication systems. Specifically, the present invention relates to joint detection of multiple user signals in a wireless communication system. [Prior Art] The first figure is a diagram of a wireless communication system. The communication system has a base station 12] to 125 which is communicable with the wireless transmission/receiving unit 7 141 141 to 143. Each base station 12] to 1L has an associated operational range, and the base station can communicate with the WTRUs 14 through 143 within its operational range. In some communication systems, such as code division multiple access (CDMA) and time division duplex (TDD/CDMA) using code division multiple f access, multiple communication messages can be transmitted on the same frequency spectrum. The type of the communication message is distinguished by the chip code (chip c〇d;,) of a plurality of communication messages. In order to use the frequency=two TDD/CDMA communication system more efficiently, communication is performed using a repeating frame divided into time slots. In this system, the message will have one or more associated codes and time slots based on the communication bandwidth. v Since multiples can be transmitted in the same spectrum and at the same time = one of the receivers must be able to retransmit the # message. It is used to detect one of the signals = system 酉 过滤 过滤 (5) price ^ ^ also ... 丨 卜 在 在 匹配 在 在 在 在 在 在 在 在 在 在 在 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 2 Ϊ Ϊ Ϊ Ϊ 可视 can be considered as interference. For Russia; multiple Us use multiple matching filters. Another method is: Disturbance Elimination (SIC). In the SIC, the communication information can be detected, and the contribution of 5 IS35735 can be deducted from the received signal for detecting the next communication message. In some cases, it is desirable to be able to detect multiple communication messages simultaneously to improve performance. The simultaneous detection of multiple communication messages is referred to as joint detection. Some joint detectors use Choiesky decomposition to implement a minimum mean square error (MMSE) detection or zero_forcing block equalizer (ZF-BLE). Other joint detection receivers use a Fourier transform based implementation to further reduce complexity.

The reason is that there is no need to implement multiple user detection in an alternative way. SUMMARY OF THE INVENTION A plurality of data signals can be received on an antenna array having a plurality of antenna elements. In a wireless communication system, the data signals can be transmitted on a spectrum. A signal having one of the data messages ^ can be received on one of the second and other components. The plurality of data signals can be grouped into a plurality of antenna elements such as a plurality of antenna elements to be matched and filtered into the first group ' to generate matching filter noises to match the filtering result to jointly detect the first courtesy point. 1 broadcast 2 with each day, line component detection data, namely -: disturbance correction signal. The result can be divided from each antenna element to produce a result for each antenna element. gp can be connected to η: interference of the antenna element. [Embodiment] The data of the remaining groups are made. Not limited to earth; I=f standby receipt: jwtru) includes but or mobile subscriber unit, 3 complaints: mobile base station, solid or % enough to operate in a wireless environment 6 1335735 Figure 2 shows the use of joint A simplified transmitter 26 and receiver 28 that combines (jd) and grouped continuous interference cancellation (GSIC), which have been used, are diversified. In a typical system, a transmitter 26 is located in the parent-WTRUs 141 through lh, and a multiple transmit circuit 26 for transmitting multiple communication messages is located at each base station 12^

in. A base station 12l will typically require at least one way 26' to communicate effectively with the WTRUs 14 through 1. The gsic_jd receiver 28 can be located at the base station 12, the WTRUs 14 to 143, and both, but the more common implementation is also based on the use of multiple components in a base 3 ί two gas. The gsic-A 28 can be transmitted from the multiple transmitter 26 or to the communication message. Shepherd

Although it has been implemented in a slotted CDMA such as TDD/CDMA, e and f CDMA (TD-CDMA)

Any other type of device. As mentioned hereinafter, a base station includes, but is not limited to, a base station, a Node B, a website controller, an access point, or other interface device in a wireless environment. ° ° = application to illustrate GSKXTD, but can also be applied to any wireless system that shares the same frequency band by multi-i communication, such as FDD/CDMA and CDMA2000. '上偻射f 2' in a wireless radio channel 3 production attack ^;. One of the emission states 26 data generator 32 4 ί on a reference channel with a receiver 28 ^ : material system according to communication bandwidth Based on the demand, one or more codes and/or time slots are widely distributed. A modulation and expansion tetgim4 is to extend the reference data, and to make a sequence of time slots and codes in the system and the appropriate allocation. 7 1335735 (time-multiplexed). In a non-grooved system, the nickname can be used for time multiplexing, such as a global pilot. The final sequence is called: $ Outbreak. The communication bursts the line frequency by a modulator 36. An antenna 38 radiates radio frequency (RF) signals to the receiving antenna array 40 via the wireless radio 30'. Modulation of transmission communication messages π ^ Any of the known phase shifting keying (DPSK), quadrature phase shift keying (qj > Sk), two amplitude modulation (QAM) known to those skilled in the art ^. In the slotted system, a typical communication burst 丨 6 with a middle code 20, Protection period (guard peH〇8, ^ data field 22, 24, as shown in Figure 3. The intermediate code two data% 22, 24 are separated, and the protection period 18 separates the communication bursts to allow transmission from different transmitters. The arrival time of the outbreak is different. The 22 and 24 sentences of the two data fields have the information of the communication outbreak, and typically have the same degree! The intermediate code 20 contains a series of sequences. The antenna array of the receiver 28 can receive various kinds of Radio frequency signal. Antenna array 40 has P antenna elements 41] to 41P. The received signal can be demodulated by demodulation transformers 42 through 42p to generate a baseband signal. A channel estimator device 44 is provided. And a J}SIC-JD device 46 is processed in the time slot, and the base frequency signal is assigned, and the appropriate code is assigned to the corresponding transmission f 26. The channel estimator device 44 The serial sequence components of the $frequency signals are used to provide channel information such as channel impulse response. The GSIC-JD device 46 will use the frequency information to estimate the transmission data of the received communication bursts as hard or soft. Figure 4. Figure 4 is a GSIC-J A simplified diagram of the D device 46. The sequence, vector, and matrix are shown in bold below, and (· indicates a complex conjugate transpose operation, and f is a transposed. K signal bursts It is valid at the same time in the same width = B. The K bursts are different codes and two, and in a UMTS TDD CDMA system, the codes include a cell-specific scrambling code and a single Or ΐ ΐ :: code. The finite transmission data code ί of length N is as in the first equation β n 4k) y, d^ev, where k = l, 2, ..., K, and the first equation The mother-poor code has a duration Γ two _ codes (all from a complex 集合 ν, two possible values, such as the second equation. , there is Λ - the first equation. According to the first code sequence ❿ are extended by the code W. W is based on (6) 4*)...4), where moxibustion = person 2, _.., especially, and "=7, 2, , 2 ^ third-party program mother code cW includes 2 cW in duration L, Each data field in each outbreak of Lizhong is filled with two pieces of ^^^. (4) Expansion factors. Although the following i 5 inch ^ ί uses a uniform spread for all κ bursts, it is easily extended by the complex number, with variable spreading factors for multiple bursts. When the data is modulated by its respective code, the radiance will typically be implemented by a transmitter (τχ) filter. The receiving antenna array has ρ antenna elements. One signal burst is through a linear independent wireless complex, and the channels have time-varying complex impulse responses (10)5, = moxibustion = 7, 2, ···, under 'and PU, ..·, / 3. Represents a transmission ^ connection with an antenna 70 pieces p. The channel output of κ bursts is based on the superposition of P receiving orders at an antenna element. Each of the superposed sequences is filtered by a receiver (RX) filter to achieve band limiting and noise suppression, and is sampled at a chip rate yA. The discrete channel pulse response h(d) for each transmitter and each antenna element can be represented by a vector f according to the fourth equation. 1^) = (0·ρ) ;^,p)...anvil, P))7·. where k=J,2,..”K,Ρ = :ΐ,2}.··,ρ,反w = 】, 2 , w The length of the fourth equation is the pulse response. Each of the samples can be obtained at the chip rate, where 妒 > 7 > ̄ e, and this method can be easily Expanded to multiple chip rate sampling. Since 妒 can be greater than &, inter-symbol interference (ISI) may occur. Typically, a reference sequence such as a midamble sequence can be used to estimate the channel impulse response h(*, P The burst response of each burst and every two antennas can be represented by the fifth equation. b(~=4f,p)硓,p)...e?),h(^)®c(〇, where k=l, 2, ':, K ' ρ = ΐ, 2, .., ρ, and 1 = 12, the length of the fifth equation symbolic response is 2+ shame / stone horse, and represents the tail of the weight Move one unit to the left side of the stone. Before processing each data field, use a +-code to eliminate B35735 2 to calculate the effect of the intermediate code on the data field. On each-piece, the length of the sequence r(8) For (#2+ team;), (p)Jn(p) II, ^ f=7 , 2,·.., household. Each—'~ is a valid sum of κ bursts and the sixth equation is not one of the m-sequences. η 矣中中1,2,...,P, and i = , (Nq+ Wj) The sixth equation of the sixth equation is the zero-mean and co-variation matrix. R(:Xp)=4(p)n(p)W],where/> = 1,2, ...,Ρ

···· ...一 .· , where corpse = person 2, ..., household seventh equation The transfer system matrix for each burst received on each antenna element is A (four) and its size is W2+. The transfer system matrix has a convolution of the transmission burst of the channel response. Each element of the transfer system matrix is in accordance with Equation 8. a(m) = (a(5,p)),

Where k = I, 2, ..., κ, p, 2, ..., P, i = J, 2, ~, 1 and Bu 1, 2, ..., J, bf'9'1 for k -

p ~ jP where I = 1,2,...,<2+w_1 n = 〇 otherwise Equation 8 Antenna p (^/β+ brother transfer system matrix eight (9) system 1335735 according to the ninth equation.

A(p) = [A0..) a{2'p) ·· A(JC-p)J

Where k = l, 2, ..., K, inverse p =], 2, :., P ninth equation The corpus of the outbreak k is the AV transfer system matrix AW according to the tenth equation.

A(i) = 'A(t.i)T A(t,2)T ··· A(i<i,)T where A:, and corpse = /,2,...,75

The tenth equation The reception sequence r(p) at the antenna p is in accordance with the eleventh equation. r(4)=(r丨(5)r2.)...also +w-i) -A^d + nip) = ^A{hp)d(i) + η(ρ) The eleventh program

The total data symbol vector is in accordance with the twelfth equation. d: W)T,··. #)Γ)Γ -[d, d2 ··· dKN) The constituent components of the twelfth equation d are in accordance with the thirteenth equation. (k) ^Ν{κ -l) + n ~ ^

Where &=i,2,··., ruler, and n = 7,2,...,iV tenth: equation gas P(NQ+W-l)xKN total transfer system matrix a is in accordance with the fourteen equation. ‘, right....le

A=“«ra (1)...A(house F)T brother fourteenth equation 12 13-35735 total noise vector n is in accordance with the fifteenth equation n=4(l)rn(2,...#)7 fifteenth The constituent components of equation n are in accordance with the sixteenth equation. 71 (WQ+1V-1)(?-〇+/ = n\ Ρ Ρ = 1,2,.··, ρ, 反 i = 】 2> ,(NQ+W 】) ώ The covariation matrix of the sixteenth equation ^the noise vector η is in accordance with the seventeenth equation - _ R„ = £'|inM}

Rp) RdX2)...R(iXiT rJx〇r(2X2) ... ♦ · · * » ♦ « * • · * ♦ rW) -^{pyj) ... where RpdpV is called, where P=U,.. ., P, and j = 1, 2, ..., P The seventeenth equation The eighteenth equation represents the total received sequence. (0r Μτ VI Γ2 ...ΓΡ("β +识-1) A d +ιι )r The constituent components of the eighteenth equation are in accordance with the nineteenth equation. r _〆r(WQ+V/-l)(/>-l)+i - ri where corpse H.,p, (corpse) anti i = 】,2,...,(NQ+Wl) 13 The nineteenth equation. The total received sequence r is in accordance with the r-h{k) +n ' t=«l

The twentieth equation rw=A(4) represents the contribution in the series. The total received signal is one of the linear equalizers of the receiving group G = preferably estimated by the use of a valued value, such as _, to determine the continuous value of the eleven equation. , (ι)τ ,... • L.,)T d: The twenty-first equation uses GSIC*^·*•, soil=== chemist, but can also use J = Ϊ ί = one forcing zero ( The ZF) criterion, and the other, uses a minimum mean square error (MMSE) criterion. The following hypothesis is that the additional noise is a random white noise with stereo space and temporary storage, and the common mutation matrix of the total noise vector is R" 〇I. (7 series of additional noise variations, and I The identity matrix of size KNxKN. With reception diversification, ZF_BLE can be derived by minimizing the quadratic cost function/(a), such as the twenty-second equation. =(r_ ^ _ The first-order equation I is a continuous fixed-valued estimate of d, and "_丨" means the inverse moment of 13-35735. The minimum value of j(aZf) will result in continuous fixed values and no offset. (unbiased) Estimate t, as in the twenty-third equation. "+ ^ A)_Vn The twenty-third equation MMSE-BLE can minimize the quadratic cost function "/(a) rot, such as the twenty-fourth equation. j(d MMSE , = £id

MMSE άϊ

MMSE d The twenty-fourth equation is a continuous fixed value estimate of a. The minimum value of the common variation matrix Rn = σ 2I by the common variation matrix of the data symbols Rd = five {ddH} = I and the total background noise vector will result in a continuous value estimation, such as the twenty-fifth equation. ^£=(Α^Α+σ2ΐΓΑ«Γ The fifteenth equation I represents the unit matrix of ΚΝχΚΝ. Since ααα is a banded bl〇ck TQepmz matrix, it is used for solution. One of the data vectors uses an approximate Cholesky formula. The Cholesky formula can be more precise and less complex, and its performance impairment is negligible. Preferably, 'to reduce complexity at the same time, and go And Multiple Access Interference (MAI), which can be combined with BLE GS GSIC (GSIC-BLE). In GSIC-BLE, it is better to collect power and divide K bursts into fewer groups. ^, with an explosion of approximately the same received power will be ^ Jun Gong: approximately the same burst has one of the combined powers received on p: equal 率 rate antenna 7L. π a Ϊ each interference cancellation stage, GSIC -BLE is old # only has 15 MAI 's of one sub-set (group) of one outbreak and jointly detects the data symbol of the group. The group code can be detected to generate MAI And it is the group that will be passed to other groups in the subsequent stages. Interference can be used In addition to removing the MAI, if the size of the group is selected to be κ, GSIC-BLE will become a single user BLE. All data can be determined in one step. ~ Fruit, the critical point of grouping can make the complexity Compromise between performances. In extreme cases, each of the κ bursts can be assigned to its own stage. This approach provides the lowest complexity. Conversely, all κ bursts can be Assigned to a single two-stage, with the highest complexity. Figure 4 is a process for receiving diverse GSIC-BLE. In GSIC_BLE with diversity of reception, all bursts S 31 are received by power or The intensity of the amplitude is sorted as; strong, as in step 5. This sorting can be achieved by previous experience at the root, or by other estimation schemes commonly used in SIC or ΐ 座 白 , , , , Channel estimation. 丄 has been ordered. Claws... 罘 十五 転 来 决定 决定 决定 决定 决定 决定 决定 决定 决定 决定 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 , the same power, that is, between each other The two explosions are divided into G groups. The order can be expressed as the force rate and in descending order. The cis/, is 1...G.%. The number of outbreaks, such as |>,=[. The receiver includes the start of the dance group 1. G stages. A joint detection for each group matrix is in accordance with the twenty-sixth equation

Where / = 7,2, ..., G MMSE-BLE second group formula twenty-seventh equation. ’

·)μλ«£ = (Α(, Α(?0 + σ21 w ) =W(i)M(i) + 1-1,2,...,0 f The temperature of the first group of the twenty-seventh equation The estimator (Wiener estimator) / = /乂...the knife will follow the twenty-eighth square

Wf=k + CT;^f A^T1 ) The twenty-eighth equation b is a unit matrix of size NxN, where N is the number of symbols in each data field of each burst.

A group BLE of the ZF-BLE The twenty-sixth equation BLE matrix is determined according to the first stage, and the transfer system matrix A(g" of the first group can be determined. A(g" is similar to the total transfer System matrix a, except that it only contains symbolic responses corresponding to bursts in the first group. In the first phase, the input sequence of group 1 can be given by the total received sequence as in the twenty-ninth The equation x(^ = r ^ ,, the ninth equation can be implemented to remove the ISI, MAI, and the near-far effect in the first group. Heavy user BLE (ZF-BLE or MMSE-BLE). In step 54, 'can take the soft decision of group 1 according to the thirtieth equation 17 (1) g, S〇ft strategy code d Λ d ω gfSqft :M? )j where JVi^), /= / 7 广++, "., G, which can be M£W or, rl). A continuous constant value estimator of the thirty-th equation, which is representative of group twist Φ 4 horse information sequence, and the code is transmitted by the first force of the first 。 ^. In step 56, can be based on the implementation of hard decision, ; = / /, hard decision 'And by: a first estimate of Yun group of the formula ^; r contribution to the p1).

• (U)T r; {itPf UX>P) The twenty-first equation ★ ., p = person 2,··.,Ρ, is the contribution of the first group to the sequence of 1 at the antenna p. In the second phase, the MAI can be eliminated by the sequence, and the rounding of the interference correction is obtained, which is obtained as the thirty-second equation. Sequence, [卜[[(2,1)Γ 1(2,2,...f f Ρ〇) ih - The twelfth equation ZF-BLE is in accordance with the thirty-third equation. Φ (0 (ΟΙαΟ'^ λ (0

A

(0H thirty-third equation MMSE-BLE φ〗 依照 according to the thirty-fourth equation 弋 18 1335735

• A thirty-fourth equation h is a unit matrix of size (NQ+JV-i)x(JVQ+W-J). The r/W is used for the antenna eve, and the new interference correction input sequence of 匕(10) is subtracted from the interference correction vector of the antenna first stage input sequence (receive sequence at the antenna p). In the subsequent step lead such as the nth phase, the new interference correction input sequence, such as the third, may be determined by subtracting the 前 group from the interference correction input sequence f in the previous stage. Fifteen equations.

ΧΪ) = χΓ) - if 】) = Κ)) χΓ) Π (1. - φ*'}) r "1 The thirty-fifth equation The product matrix is in accordance with the thirty-sixth equation. If a &b; 搞, if a &b;

The thirty-sixth equation is similar to the first stage, including 乂(4), and it is possible to implement a single user or multiple users BLe j to remove the 6 φ group, and the problem of its own closeness. In the step code. In the 'can' thirty-seventh equation to represent the soft decision symbol twenty-seven equation in step 62, you can use the soft decision character: 2 hard decision to generate a hard decision character to drink a (L: + day by ^, the The hard code is used to generate the first contribution in r, such as the thirty-eighth equation. Group, > and the shell 19 丄 335735 = A? C flute ~ upper, brother twenty-eight equation (four) is similar to the first stage Including for each antenna *, household =; 乂 ..., household. In the first stage, can be i έ ;; = lose j column ί in addition to the ΜΑΙ to get the interference repair is entering the input 歹 J ' as the thirtieth Nine equations and step 66. ί (0 program thirty-ninth equation in the last step, the input sequence will become the fortieth party 4C) = 4C-1) - rf fortieth equation = Π (Ir <) The implementation of a single or multiple user BLE, that is, the forty-first equation, leads to the decision code. a C^=Mf)xf) The fourth + singer's forty-one formula can use hard decision making, and the hard decision code from the soft decision to the final stage. By considering the segments as linear filtering of one of the received sequences, each linear filter 卜, 卜 7 G, will follow the fourth program. Μ ΠΚ ΠΚ) Μ 0) Λ Program Forty-second square The soft decision code for each stage will follow the fourth • _ I »- square 20 A (0

i/>A =diag(^tnH A )d + dSJ^(〇wa ]d ma 严+ 4,)n The H(x)i generation of the forty-third equation _! The matrix x contains only diagonals The element "pair: matrix. (X) represents a matrix of zero-diagonal elements containing only the elements other than X. ★ In the three programs, the first item represents *, the group μ匕 symbol, the second item represents the first group, the group ISIJ = item, and the last-item is the background of the z-stage output. Line 3; 2f a vector ' * the first y component ί ί ί ΐ ΐ f rigorous code vector. The seventh component of the sub-segment data is stepped on a β 2 a, the j-th, and the component quantity is in the total transmission. All the other transmission codes are - Ξ < correlation is the co-mutation moment ^, ', 口疋, where Rn is the total receiving sequence (4) the total number of variances in the winter stage output = ^ ^ 唬 interference and noise ratio (SIN R) according to the forty-fourth ^ where FWWA, j = n + N ( Kl),i=l,2,...,G, H'2 μ,2,..,ν Forty-fourth equation 4 engraving 并i and recognizing-*summer LxJj,j system is not matrix X Column 7 of the 7th prime. ;r = to , , 7 in the number amp & 丄 x叩u < Dan variation matrix. In the numerical simulation, full BLE_FBLE (with only a single 21 1335735 stage BLE) shows better performance than GSIC-BLE. When considering the encoding gain of 1% to 10% uncoded bit error rate (BER), the performance of GSIC-BLE will be close to that of FBLE. GSIC-BLE is also suitable for multiple code schemes where some or all of the users transmit multiple codes. Multiple codes from the same user can be grouped together and multiple user BLEs can be implemented on each group. The MAI between these groups will be eliminated by the SIC. GSIC-BLE will win the performance of better than the traditional SIC in two ways. First, unlike the conventional SIC, it can maintain its performance in the absence of a near-far effect by implementing a multi-user BLE with an explosion of similar received power. Second, unlike the RAKE-based SIC receivers in the prior art, it is possible to handle the ISI of each burst perfectly through the multi-user BLE of each group. Optimizing ISI mitigation will effectively eliminate MAI between multiple groups, especially in channels with large delay propagation. GISC-BLE typically achieves a complexity that varies linearly with the number of bursts K, but is generally lower than that of FBLE. Since this situation can handle the ISI in each burst, it is possible to achieve better performance than a RAKE-based SIC receiver. This performance advantage is even stronger in channels with large delay propagation, ie when the ISI is severe. Even for channels with large propagation delays, the near-intensity effect of the order of 2 decibels (dB) between multiple bursts still shows the effectiveness of those who are comparable to FBLE. Figure 5 is a simplified block diagram of GSIC-BLE combined with one of the diverse uses of reception. Receive vectors from P antenna elements? / Call to will be input to GSIC-BLE. A group 1 matching filter 70 performs matching on the reception vector of group 1 22 B35735 Filter «. One of the results of the matching filter is processed by a BLE, such as by a ZF (« 彳, or MMSE («付 2/)-丨乂丨). One result (10) of BLE 72 will be converted to a hard symbol-to-interference by a soft-to-hard decision-making device 74, which uses the hard-coded code to generate a vector f to one for each antenna, Represents the contribution of group i to the antenna receive vector. For each antenna, a subtractor 92] to 92P will derive the self-receiving vector to the deduction of the group j to /^> to generate an interference cancellation vector ^ ' to f/&quot for each antenna. ;". The Group 2 Matching Filter 78 performs matching filtering on the interference cancellation vectors 42)%?). One of the matching filters results # is processed by a BLE 80, such as (<)~/)->) when a ZF, or («+σ2/)^ when MMSE. One of the BLEs? ▲ /2) is converted into a hard code by a soft to hard decision device 82. The interference correction device 84 uses the hard symbols to generate a vector for each antenna. Ρ η, whose table group 2 contributes s to the antenna reception vector. For every two antennas, a subtractor 9^ to 94P will be wide from the receive vector to the wide] except for the contribution of group 2 to f,) to generate a pair of interference cancellation vectors Fg( , ) to & Ming. The parent antenna battalion two If, the remaining group ‘, ie, group 3 to group W Ϊ is eliminated until the final group G is the group G matching filter, the 8.6 pair is dry, and the elimination vector implements matching filtering. One of the matching filters is handled by a BLE 88, such as by one; m, or MMSE. B] l = a result < 2 will be converted from a soft to hard ^ into a hard code to the 决 table device 90 and 23 13.35735 [Simplified illustration] Figure 1 is a simplified diagram of a wireless communication system formula. Figure 2 is a simplified block diagram of a transmitter and a combined continuous interference cancellation receiver with multiple antenna elements. Figure 3 is a diagram of a communication outbreak (outbreak). Figure 4 is a flow chart for continuous interference cancellation for joint detection groups with multiple antenna elements. Figure 5 is a simplified block diagram of a joint detection group continuous interference canceller.

18 protection period 22 data field 26 transmitter 32 data generator 36 modulator 42, 42p demodulator 44 channel estimator device [main component symbol description] 16 communication burst 20 intermediate code 24 data field 28 receiver 34 modulation And expansion device 30 wireless radio channel 46 GSIC joint detection device

twenty four

Claims (1)

1335735 ΛΤ y X. Patent application scope: L + I receivers are used to receive a plurality of radio frequency (RJF) lean signals on a dry spectrum, the receiver comprising: an antenna array having a plurality of antenna elements; Demodulation. Cry, jL = change to the complex data signal received on each antenna element, ^2 demodulation = corresponds to the "multiple antenna elements", wherein each of the demodulation transformers is adapted The plurality of antenna elements of the plurality of antenna elements are received to generate a baseband signal; a channel estimation device configured to provide channel information; and - group continuous interference cancellation (GSIC) joint detection ( a JD) device configured to: divide the complex data signal into a plurality of groups; receive a credit to scream the received group of the plurality of groups _--the group towel antenna element to generate a matched filtering result; Data. Use the matching filter and the wave result to jointly detect the number corresponding to the first group; the data construction side of each antenna element-interference correction signal (4) u receive the money from each antenna element and subtract the interference of the component Positive a produces an interference cancellation result corresponding to each antenna element; and, the target ancestor a ^ is continuously detected by the interference cancellation result corresponding to each antenna element (10), as the material signal is the hard group of the remaining group Evaluation of soft codes. 2. The receiver of claim 1 of the patent application, which provides the parental information based on a series of phase components of the base age towel. °...again 3 - kind of transmitter 'set to money in a shared frequency (RF) data, the transcoder contains: job radio frequency 25 1335735 °d Λ η; data signal generator '(four) generated will be in - Referring to the reference 2 transformer of the peer communication, the rib transforms the communication burst into a radio frequency (RF) antenna for radiating the data signal to a receiver by a wireless radio channel. ^ The transmitter of claim 3, wherein the extension is arranged with the modulation device to generate a communication burst, the communication bursting into an almost continuous full area, code. 5. The transmitter of claim 3, wherein the modulator is configured to use a differential phase shift keying (DPSK). 6. The transmitter of claim 3, wherein the modulator is configured to use a quadrature phase shift keying (QPSK). 7. The transmitter of claim 3, wherein the modulator is configured to use a quadrature amplitude modulation (QAM). 8. The transmitter of claim 3, wherein the expansion and modulation device is configured to generate a communication burst comprising: a midamble, a guard period, and two data fields. 9. The device of claim 8, wherein the expansion and modulation device is arranged to generate a communication burst between the two fields. 10. The transmitter of claim 8 wherein the expansion and modulation means are arranged to generate a communication burst, wherein the guard period separates the communication burst from the next communication burst. 11. The transmitter of claim 8 wherein the expansion and modulation means are arranged to produce a communication burst of the same length of the two data fields. 26